The study presents a model in which two processes are hypothesized to influence the relationships between pre-teenage children's exposure to community violence and their early adolescent outcomes--especially whether they become victims or perpetrators of violence. One process addresses variables influencing the amount of violence the children are personally exposed to. The second process addresses variables influencing the effects on outcome of the amount of violence the children personally experience. From a pool of 2,100 6th graders living in a high-crime urban neighborhood, 600 children--200 of whom are rated by their classmate peers as aggressive, 200 as teased or picked on, and 200 as cooperative--will be followed for three years. Each year the investigators will assess both their exposure to community violence and outcome. In the first two years, we will assess the modifying factors hypothesized to protect against poor outcome. Interviews with the children will provide information on exposure to community violence. Interviews with children and parents will provide information on the modifying factors in the preceding year(s). Outcome will be assessed in terms of whether the children become perpetrators of violent behavior; whether they become victims of violent behavior; the extent to which they display internalizing problems or distress symptoms. The indirect effect of exposure to violence on academic achievement through involvement and poor mental health will also be evaluated.